# RLinf: Flexible and Efficient Large-scale Reinforcement Learning via Macro-to-Micro Flow Transformation

arXiv: https://arxiv.org/abs/2509.15965 · HTML: https://mlsanity.com/paper/2509.15965/

## Key facts

- Cited by 57 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 31 of its 54 references that are in the map.
- First posted on arXiv 19 Sep 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 30 of 31)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://mlsanity.com/paper/1909.08053/) (2019)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success](https://mlsanity.com/paper/2502.19645/) (2025)
- [Training Deep Nets with Sublinear Memory Cost](https://mlsanity.com/paper/1604.06174/) (2016)
- [PyTorch FSDP: Experiences on Scaling Fully Sharded Data Parallel](https://mlsanity.com/paper/2304.11277/) (2023)
- [Ray: A Distributed Framework for Emerging AI Applications](https://mlsanity.com/paper/1712.05889/) (2017)
- [ZeRO-Offload: Democratizing Billion-Scale Model Training](https://mlsanity.com/paper/2101.06840/) (2021)
- [A Survey of Embodied AI: From Simulators to Research Tasks](https://mlsanity.com/paper/2103.04918/) (2021)
- [ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning](https://mlsanity.com/paper/2104.07857/) (2021)
- [AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning](https://mlsanity.com/paper/2505.24298/) (2025)
- [ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI](https://mlsanity.com/paper/2410.00425/) (2024)
- [DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales](https://mlsanity.com/paper/2308.01320/) (2023)
- [What Can RL Bring to VLA Generalization? An Empirical Study](https://mlsanity.com/paper/2505.19789/) (2025)
- [Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training](https://mlsanity.com/paper/2110.14883/) (2021)
- [AceMath: Advancing Frontier Math Reasoning with Post-Training and Reward Modeling](https://mlsanity.com/paper/2412.15084/) (2024)
- [NeMo-Aligner: Scalable Toolkit for Efficient Model Alignment](https://mlsanity.com/paper/2405.01481/) (2024)
- [FastCuRL: Curriculum Reinforcement Learning with Stage-wise Context Scaling for Efficient Training R1-like Reasoning Models](https://mlsanity.com/paper/2503.17287/) (2025)
- [Priority-based Parameter Propagation for Distributed DNN Training](https://mlsanity.com/paper/1905.03960/) (2019)

## Built on by (most cited first, 30 of 57)

- [WoVR: World Models as Reliable Simulators for Post-Training VLA Policies with RL](https://mlsanity.com/paper/2602.13977/) (2026)
- [Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agents](https://mlsanity.com/paper/2607.08448/) (2026)
- [Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments](https://mlsanity.com/paper/2605.30280/) (2026)
- [G0.5: One Autoregressive Stream for Robot Reasoning and Action](https://mlsanity.com/paper/2608.11739/) (2026)
- [AsyncVLA: Asynchronous Flow Matching for Vision-Language-Action Models](https://mlsanity.com/paper/2511.14148/) (2025)
- [Simple Recipe Works: Vision-Language-Action Models are Natural Continual Learners with Reinforcement Learning](https://mlsanity.com/paper/2603.11653/) (2026)
- [π_textttRL: Online RL Fine-tuning for Flow-based Vision-Language-Action Models](https://mlsanity.com/paper/2510.25889/) (2025)
- [DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training](https://mlsanity.com/paper/2604.26256/) (2026)
- [ProgressVLA: Progress-Guided Diffusion Policy for Vision-Language Robotic Manipulation](https://mlsanity.com/paper/2603.27670/) (2026)
- [Beyond Imitation: Reinforcement Learning-Based Sim-Real Co-Training for VLA Models](https://mlsanity.com/paper/2602.12628/) (2026)
- [ProcVLM: Learning Procedure-Grounded Progress Rewards for Robotic Manipulation](https://mlsanity.com/paper/2605.08774/) (2026)
- [NoiseGate: Learning Per-Latent Timestep Schedules as Information Gating in World Action Models](https://mlsanity.com/paper/2605.07794/) (2026)
- [π-StepNFT: Wider Space Needs Finer Steps in Online RL for Flow-based VLAs](https://mlsanity.com/paper/2603.02083/) (2026)
- [IG-RFT: An Interaction-Guided RL Framework for VLA Models in Long-Horizon Robotic Manipulation](https://mlsanity.com/paper/2602.20715/) (2026)
- [TempoFit: Plug-and-Play Layer-Wise Temporal KV Memory for Long-Horizon Vision-Language-Action Manipulation](https://mlsanity.com/paper/2603.07647/) (2026)
- [Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts](https://mlsanity.com/paper/2605.22446/) (2026)
- [Don't Drop the BATON: Long-Horizon Robot Manipulation via Agentic Subtask Exploration and Transition-aware Memory](https://mlsanity.com/paper/2608.16889/) (2026)
- [Sword: Style-Robust World Models as Simulators via Dynamic Latent Bootstrapping for VLA Policy Post-Training](https://mlsanity.com/paper/2605.07288/) (2026)
- [Libra: Efficient Resource Management for Agentic RL Post-Training](https://mlsanity.com/paper/2606.03077/) (2026)
- [RedFlow: Redirect Failure into Action-Level Corrections for Flow-matching VLA Policy](https://mlsanity.com/paper/2607.27782/) (2026)
- [Thousand-GPU Large-Scale Training and Optimization Recipe for AI-Native Cloud Embodied Intelligence Infrastructure](https://mlsanity.com/paper/2603.11101/) (2026)
- [ActProbe: Action-Space Probe for Early Failure Detection of Generative Robot Policies](https://mlsanity.com/paper/2606.08508/) (2026)
- [TacCoRL: Integrating Tactile Feedback into VLA via Simulation](https://mlsanity.com/paper/2606.11743/) (2026)
- [FBFM: A Training-Free Asynchronous Feedback Mechanism for Flow-Matching in World-Action Models Execution](https://mlsanity.com/paper/2607.29235/) (2026)
- [Continue or Replan? Bernoulli-Continuation Policy Learning for Adaptive Horizon Execution](https://mlsanity.com/paper/2608.03483/) (2026)
- [FiberTune: Preserving Action-Fiber Visual Residuals in Vision-Language-Action Fine-Tuning](https://mlsanity.com/paper/2606.08653/) (2026)
- [ReCoVLA: VLM-Guided Reward Compilation for Failure Recovery in Vision-Language-Action Policies](https://mlsanity.com/paper/2606.09630/) (2026)
- [PolicyTrim: Boosting Intrinsic Policy Efficiency of Vision-Language-Action Models](https://mlsanity.com/paper/2606.22540/) (2026)
- [DenseReward: Dense Reward Learning via Failure Synthesis for Robotic Manipulation](https://mlsanity.com/paper/2607.13033/) (2026)
- [DynaResize: Runtime GPU Reallocation for Disaggregated LLM Post-Training](https://mlsanity.com/paper/2607.22614/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) has demonstrated immense potential in advancing artificial general intelligence, agentic intelligence, and embodied intelligence. However, the inherent heterogeneity and dynamicity of RL workflows often lead to low hardware utilization and slow training on existing systems. In this paper, we present RLinf, a high-performance RL training system based on our key observation that the major roadblock to efficient RL training lies in system flexibility. To maximize flexibility and efficiency, RLinf is built atop a novel RL system design paradigm called macro-to-micro flow transformation (M2Flow), which automatically breaks down high-level, easy-to-compose RL workflows at both the temporal and spatial dimensions, and recomposes them into optimized execution flows. Supported by RLinf worker's adaptive communication capability, we devise context switching and elastic pipelining to realize M2Flow transformation, and a profiling-guided scheduling policy to generate optimal execution plans. Extensive evaluations on both reasoning RL and embodied RL tasks demonstrate that RLinf consistently outperforms state-of-the-art systems, achieving 1.07×-2.43× speedup in end-to-end training throughput.